Compensation type circuit breaker trip unit based on TS type fuzzy controller

Abstract

The present invention discloses a compensating circuit breaker trip unit based on a T-S type fuzzy controller, which relates to the field of power electronics technology. The compensating circuit breaker trip unit includes an actuator, a compensation circuit, and a fuzzy controller. The actuator is an electromagnet for converting electromagnetic energy into mechanical energy to perform a circuit breaker tripping operation. The compensation circuit is connected to the electromagnet and is used to adjust the magnitude of the compensation current according to a control signal to adjust the electromagnetic attraction of the electromagnet. The fuzzy controller is a T-S type fuzzy controller for receiving a current signal from a circuit where the circuit breaker is located and outputting a control signal according to preset fuzzy rules and a T-S model to control the working state of the compensation circuit, thereby preventing the trip unit from malfunctioning or refusing to operate. The present invention uses a T-S type fuzzy controller to control the current magnitude of the compensation circuit of the circuit breaker, which can effectively prevent the trip unit from malfunctioning or refusing to operate, and prevent the short-circuit protection function of the circuit breaker from failing.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronics, and in particular to a compensating circuit breaker release based on a TS-type fuzzy controller. Background Art

[0002] The electromagnet is the actuator of the short-circuit release, an electromagnetic element that converts electromagnetic energy into mechanical energy. When a short circuit occurs in the circuit, the short-circuit current flows through the release coil, generating an attractive force on the iron core, which pulls the drawbar, unlocking the latch connected to the drawbar and breaking the circuit. Short-circuit protection failure modes include malfunction and refusal of the release.

[0003] The reasons for malfunction are: (1) attenuation of the electromagnet reaction spring force. If the electromagnetic attraction remains unchanged, the reaction spring force is insufficient. The electromagnetic attraction under the non-operating current overcomes the spring resistance and pulls the traction rod, causing the release to malfunction. (2) Excessive electromagnetic attraction. The internal structure of the miniature circuit breaker is compact. Under high current, the temperature rises, the iron core expands, and the cross-sectional area increases. The electromagnetic attraction increases, and after overcoming the spring resistance, it pulls the traction rod, causing the release to malfunction.

[0004] The reasons for refusal to operate are: (1) The distance between the electromagnet core and the traction rod is too large. Under the fault current, when the electromagnetic attraction causes the core to move to its maximum stroke, there is not enough pulling force to pull the traction rod. In most cases, the traction rod is only fixed by a thin spring, which is not firm. (2) Increased current loss. In addition to the coil resistance loss, the coil current also has hysteresis and eddy current losses in the magnetic conductor. As a result, the electromagnetic attraction is insufficient to pull the traction rod under the fault current.

[0005] The existing technology presents the following challenges: A compensation circuit can effectively address the issue of trip unit malfunction and rejection. When the reaction spring force within the electromagnet decays, the compensation current decreases, reducing the electromagnetic attraction and preventing malfunction. However, when the distance between the ferromagnetic core and the drawbar is too large, or when current loss increases, the compensation current increases, increasing the electromagnetic attraction and preventing trip unit rejection. However, controlling the compensation current remains a pressing challenge. Summary of the Invention

[0006] The present invention aims to provide a compensating circuit breaker release based on a TS-type fuzzy controller to solve the problems raised in the above background technology.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] A compensating circuit breaker tripper based on a TS-type fuzzy controller, the compensating circuit breaker tripper comprising an execution component, a compensation circuit and a fuzzy controller;

[0009] Wherein, the actuator is an electromagnet, which is used to convert electromagnetic energy into mechanical energy to perform the tripping operation of the circuit breaker;

[0010] The compensation circuit is connected to the electromagnet and is used to adjust the magnitude of the compensation current according to the control signal to adjust the electromagnetic attraction of the electromagnet;

[0011] The fuzzy controller is a TS-type fuzzy controller, which is used to receive the current signal of the circuit breaker and output a control signal according to preset fuzzy rules and the TS model to control the working state of the compensation circuit, thereby preventing the trip unit from malfunctioning or refusing to operate.

[0012] A further improvement of the technical solution of the present invention is that the TS-type fuzzy controller is a single-input-single-output system, and its control process includes the following steps:

[0013] The current of the circuit breaker is monitored in real time, and a clear input quantity i is obtained through a current transformer. The clear input quantity i is the current value in the circuit breaker;

[0014] The acquired current data is input into the TS-type fuzzy controller for structure identification and parameter identification;

[0015] According to the identification result, a clear output quantity u is calculated and outputted through a preset fuzzy rule. The clear output quantity u is a compensation current value signal. The calculated compensation current value signal is then outputted to the compensation circuit.

[0016] The compensation circuit receives the compensation current value signal from the TS-type fuzzy controller, and outputs an equivalent compensation current according to the compensation current value signal. By adjusting the magnitude of the compensation current, the electromagnetic attraction of the electromagnet is dynamically adjusted to ensure that the electromagnetic attraction is appropriately reduced when the elastic force of the reaction spring in the electromagnet decays, thereby preventing the release from malfunctioning. When the overlap distance between the iron core of the ferromagnet and the traction rod is too large or the current loss increases, the electromagnetic attraction can be increased accordingly to avoid the release from refusing to operate, thereby effectively ensuring that the short-circuit protection function of the circuit breaker works normally and reliably.

[0017] A further improvement of the technical solution of the present invention is that the structure identification and parameter identification specifically include:

[0018] Perform structural identification on the data of the clear input quantity i of the TS-type fuzzy controller to determine whether to use the 0th-order or 1st-order linear model;

[0019] After determining the form of the linear model, further parameter identification is performed on the input current data based on the selected linear model to determine the parameters a and k in the model. For the 0th-order model, the goal of parameter identification is to determine the constant k so that it can accurately reflect the output level of the system under steady-state conditions. For the 1st-order model, constants a and k need to be determined simultaneously to describe the linear relationship between the input current and the output.

[0020] According to the results of structure identification and parameter identification, a complete TS model is established, and a clear output quantity u is output according to the model.

[0021] A further improvement of the technical solution of the present invention is that the control rules of the TS-type fuzzy controller specifically include:

[0022] Define the fuzzy set A, and determine the corresponding fuzzy rule according to the fuzzy set A to which the clear input quantity i belongs;

[0023] In the 0th order model, the fuzzy rule is "if the clear input quantity i belongs to A, then u = k", where k is a constant related to the set A. The 0th order model is suitable for situations where the relationship between the system output and input is relatively simple, and the system is directly controlled by the constant output;

[0024] In the first-order model, the fuzzy rule is "if the clear input quantity i belongs to A, then u = ai + k", where a and k are constants related to the set A. The first-order model can describe the linear relationship between input and output and is suitable for situations where the system dynamic characteristics are relatively complex. The output is adjusted in the form of a linear function.

[0025] The corresponding fuzzy rules are activated according to the membership degree of the clear input quantity i, and the clear output quantity u is calculated according to the rules.

[0026] A further improvement of the technical solution of the present invention is that the working mode of the compensation circuit specifically includes:

[0027] When the reaction spring force inside the electromagnet decays, the compensation current calculated according to the TS model decreases, and the electromagnetic attraction decreases accordingly to prevent the release from malfunctioning.

[0028] When the distance between the ferromagnetic core of the electromagnet and the traction rod is too large, or the current loss increases, the compensation current calculated according to the TS model increases, and the electromagnetic attraction increases accordingly to avoid the release from refusing to operate;

[0029] The compensation circuit dynamically balances the electromagnetic attraction of the electromagnet and the elastic force of the reaction spring by adjusting the magnitude of the compensation current;

[0030] The compensation circuit outputs a compensation current of equal value according to the received compensation current value signal, thereby accurately adjusting the electromagnetic attraction force of the electromagnet.

[0031] A further improvement of the technical solution of the present invention is that the output of the TS-type fuzzy controller specifically includes:

[0032] When the clear input quantity i is input into the TS-type fuzzy controller, the corresponding fuzzy rules are activated according to their membership, and the weight coefficient w of each rule is determined. Each activated fuzzy rule corresponds to a weight coefficient w, which reflects the degree to which the clear input quantity i belongs to the fuzzy set corresponding to the fuzzy rule.

[0033] The TS-type fuzzy controller uses the weighted summation method or weighted average method to calculate the final output U;

[0034] In the weighted summation method, the final output U is obtained by multiplying the clear output u of all activated fuzzy rules with their corresponding weight coefficients w and then summing them up. The calculation formula is: ;

[0035] The final output U is obtained by multiplying the sum of the clear output u of all activated fuzzy rules with their corresponding weight coefficients w, divided by the sum of all weight coefficients. The calculation formula is: , where u is the output of each rule and w is the corresponding weight coefficient.

[0036] A further improvement of the technical solution of the present invention is that the determination of the weight coefficient of each rule specifically includes:

[0037] The minimum method is used to determine the weight coefficient, that is, the minimum membership degree of each fuzzy rule is taken as the weight coefficient;

[0038] The weight coefficient is determined by the product method, that is, the product of the membership degree of each fuzzy rule is taken as the weight coefficient;

[0039] According to the actual application scenario and control requirements, the weight coefficient is manually determined through human experience or experimental data to optimize the control effect.

[0040] A further improvement of the technical solution of the present invention is that the compensation circuit further adopts a compensation circuit based on a neural network, and its working method includes:

[0041] The compensation current value signal from the fuzzy controller is received, and the compensation current is adaptively adjusted through a compensation model constructed based on a neural network. The training data of the neural network includes the current signal and the corresponding compensation current value under different working conditions to control the electromagnetic attraction of the electromagnet.

[0042] A further improvement of the technical solution of the present invention is that the construction process of the compensation model is:

[0043] The current signals and corresponding compensation current values ​​under different operating conditions are collected, covering various situations such as normal working conditions, reaction spring force attenuation, excessive overlap between the core and the drawbar, and increased current loss. The collected data is normalized, and the current signals and compensation current values ​​are mapped to the same range. The normalized data is then integrated and divided into training and test sets to design the neural network structure, including the input layer, hidden layer, and output layer.

[0044] A compensation model is constructed based on the designed neural network structure. The training set is used as input into the neural network structure for training. The weights and biases of the neural network are randomly initialized. The input current signal is forward-propagated through the neural network, and the output compensation current value is calculated. The trained neural network is then verified using an independent test data set. The mean square error index on the test data is calculated to evaluate the performance of the neural network. The neural network is then optimized to ultimately obtain the constructed compensation model.

[0045] The constructed compensation model is deployed and the compensation circuit is used to receive the compensation current value signal from the TS-type fuzzy controller. The input signal is input into the compensation model, and the final compensation current value is calculated through forward propagation. According to the output of the compensation model, the compensation circuit outputs the corresponding compensation current to adjust the electromagnetic attraction of the electromagnet.

[0046] A further improvement of the technical solution of the present invention is that the TS-type fuzzy controller further includes a fault diagnosis module for real-time monitoring of the current signal of the circuit breaker and the working status of the compensation circuit. When an abnormal signal is detected, the fault diagnosis module issues an alarm signal and transmits the abnormal information to an external monitoring system for fault detection and processing.

[0047] Among them, the specific workflow of the fault diagnosis module is as follows:

[0048] The fault diagnosis module continuously monitors the current signal of the circuit breaker and the working status of the compensation circuit in real time, obtains the real-time value of the line current, and monitors the current, voltage and electromagnetic attraction data of the electromagnet in the compensation circuit;

[0049] The collected signals and data are analyzed and processed in real time, and the presence of abnormal signals is detected through preset threshold judgment. When the line current exceeds the set normal range threshold, or the parameters of the compensation circuit show changes that do not conform to the normal working mode, it is identified as an abnormal signal;

[0050] Once an abnormal signal is detected, the fault diagnosis module immediately triggers the alarm mechanism and sends out an alarm signal to alert on-site personnel. At the same time, the abnormal information including the specific parameters of the abnormal signal, the time of occurrence and the cause is transmitted to the external monitoring system through the communication interface. After receiving the abnormal information, the external monitoring system quickly locates the fault location and provides accurate fault information to maintenance personnel so that they can conduct timely fault investigation and processing, thereby minimizing the impact of the fault on the operation of the power system and ensuring the safe and stable operation of the power system.

[0051] Due to the adoption of the above-mentioned technical solution, the present invention achieves the following technical advances compared to the prior art: the use of a compensation circuit can effectively solve the problem of malfunction and refusal of the release, and the TS-type fuzzy inference output can be directly used to control the compensation circuit, capable of approximating any nonlinear system; the fuzzy controller is a single-input-single-output system, which establishes a TS model by performing structural and parameter identification on current data, and the TS model determines the operating mode of the compensation circuit: when the elastic force of the reaction spring inside the electromagnet decays, the compensation current decreases, and the electromagnetic attraction decreases, preventing malfunction of the release; when the overlap distance between the ferromagnetic core and the draw rod is too large, or the current loss increases, the compensation current increases, and the electromagnetic attraction increases, preventing malfunction of the release. Therefore, the use of a TS-type fuzzy controller to control the current in the circuit breaker's compensation circuit can effectively prevent malfunction and refusal of the release, and prevent the short-circuit protection function of the circuit breaker from failing. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0053] Figure 1 A working principle diagram of a compensating circuit breaker release based on a TS-type fuzzy controller provided by the present invention;

[0054] Figure 2 A diagram showing the relationship between the clear input i and the clear output u of a single-input-single-output system of a TS-type fuzzy controller provided by an embodiment of the present invention;

[0055] Figure 3 The position of the input quantity of the TS-type fuzzy controller provided by the embodiment of the present invention in the fuzzy subset mf1;

[0056] Figure 4 The position of the input quantity of the TS-type fuzzy controller provided by the embodiment of the present invention in the fuzzy subset mf2;

[0057] Figure 5 The position of the input quantity of the TS-type fuzzy controller provided by the embodiment of the present invention in the fuzzy subset mf3. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0059] Example 1, as Figure 1 As shown, the present invention provides a compensating circuit breaker release based on a TS-type fuzzy controller, the compensating circuit breaker release comprising an execution component, a compensation circuit and a fuzzy controller;

[0060] The actuator is an electromagnet, which is used to convert electromagnetic energy into mechanical energy to perform the tripping operation of the circuit breaker;

[0061] The compensation circuit is connected to the electromagnet and is used to adjust the magnitude of the compensation current according to the control signal to adjust the electromagnetic attraction of the electromagnet;

[0062] The fuzzy controller is a TS-type fuzzy controller that receives the current signal of the circuit breaker and outputs a control signal based on the preset fuzzy rules and TS model to control the working state of the compensation circuit, thereby preventing the trip unit from malfunctioning or refusing to operate.

[0063] The TS fuzzy controller is a single-input-single-output system, and its control process includes the following steps:

[0064] The current of the circuit where the circuit breaker is located is monitored in real time, and a clear input quantity i is obtained through a current transformer. The clear input quantity i is the current value in the circuit where the circuit breaker is located. The obtained current data is input into a TS-type fuzzy controller for structure identification and parameter identification. According to the identification result, a clear output quantity u is calculated and output through a preset fuzzy rule. The clear output quantity u is a compensation current value signal. The calculated compensation current value signal is then output to a compensation circuit. The compensation circuit receives the compensation current value signal from the TS-type fuzzy controller and outputs an equivalent compensation current according to the compensation current value signal. By adjusting the magnitude of the compensation current, the electromagnetic attraction of the electromagnet is dynamically adjusted to ensure that the electromagnetic attraction is appropriately reduced when the elastic force of the reaction spring in the electromagnet decays, thereby preventing the release from malfunctioning. When the overlap distance between the iron core of the ferromagnet and the traction rod is too large or the current loss increases, the electromagnetic attraction can be increased accordingly to prevent the release from refusing to operate, thereby effectively ensuring that the short-circuit protection function of the circuit breaker works normally and reliably.

[0065] Structural identification and parameter identification specifically include:

[0066] The data of the clear input quantity i of the TS-type fuzzy controller is structurally identified to determine whether to use a 0th-order or 1st-order linear model. The TS-type fuzzy controller first analyzes the characteristics of the input data to determine the most suitable linear model form for describing the input-output relationship of the system. The core of structural identification is to use a data-driven method to determine whether the system is more suitable for a 0th-order linear model (i.e., a constant model) or a 1st-order linear model (i.e., a linear function model). After determining the form of the linear model, the input current data is further parameterized according to the selected linear model to determine the parameters a and k in the model. For the 0th-order model, the goal of parameter identification is to determine the constant k so that it can accurately reflect the output level of the system under steady-state conditions. For the 1st-order model, the constant a needs to be determined at the same time. and k, to describe the linear relationship between input current and output. According to the results of structure identification and parameter identification, a complete TS model is established, and a clear output quantity u is output according to the model. The relationship between the clear input quantity i and the output clear output quantity u is expressed in the form of a mathematical formula through the TS model. Based on the established TS model, the TS-type fuzzy controller can quickly and accurately output the corresponding clear output quantity u according to the real-time input clear input quantity i through fuzzy reasoning and calculation. The clear output quantity u is used as the compensation current value to directly act on the compensation circuit to guide it to adjust the size of the compensation current, thereby realizing dynamic adjustment of the electromagnetic attraction of the electromagnet, ensuring that the circuit breaker release can work stably and reliably under various complex working conditions, and effectively preventing the occurrence of false operation and refusal to operate.

[0067] The control rules of the TS fuzzy controller specifically include:

[0068] Define a fuzzy set A, and determine the corresponding fuzzy rule based on the fuzzy set A to which the clear input quantity i belongs. The fuzzy set is used to fuzzify the clear input quantity i. Each fuzzy set A corresponds to a membership function, which is used to describe the degree to which the clear input quantity i belongs to the fuzzy set. According to the value of the clear input quantity i, determine the fuzzy set A to which it belongs, and match the corresponding fuzzy rule. In the 0th order model, the fuzzy rule is "If the clear input quantity i belongs to A, then u=k", where k is a constant related to the set A. The 0th order model is suitable for situations where the relationship between the system output and input is relatively simple, and the system is directly controlled by the constant output. In the 1st order model, the fuzzy rule is "If the clear input quantity i belongs to A, then u=ai+k ", where a and k are constants related to the set A. The first-order model can describe the linear relationship between input and output and is suitable for situations where the system dynamic characteristics are relatively complex. The output is adjusted in the form of a linear function. The corresponding fuzzy rules are activated according to the membership of the clear input i, and the clear output u is calculated according to the rules. Since the clear input i may belong to multiple fuzzy sets at the same time, multiple fuzzy rules may be activated. Each activated fuzzy rule will calculate a local clear output u according to its corresponding fuzzy rule form. In addition, according to the fuzzy rules, n control rules can be used to describe the system. When the specific input data is x, m control rules will be activated, and the final output U will be determined by the clear output u of these m control rules.

[0069] The working mode of the compensation circuit specifically includes:

[0070] When the elastic force of the reaction spring in the electromagnet decays, the compensation current calculated according to the TS model decreases, and the electromagnetic attraction decreases accordingly to prevent the release from malfunctioning. When the reaction spring in the electromagnet decays due to long-term use or environmental factors, the electromagnetic attraction of the electromagnet will be relatively enhanced under the same current. At this time, the compensation current value calculated according to the TS model will decrease accordingly. By reducing the compensation current, the electromagnetic attraction of the electromagnet will also decrease, thereby avoiding the release from malfunctioning due to excessive electromagnetic attraction under normal operating current. This compensation mechanism can effectively extend the service life of the release and ensure that it can still work normally when the reaction spring ages. When the overlap distance between the iron core of the ferromagnet in the electromagnet and the traction rod is too large, or the current loss increases, the compensation current calculated according to the TS model increases, and the electromagnetic attraction increases accordingly to avoid the release from refusing to operate. When the overlap distance between the iron core of the electromagnet and the traction rod increases due to mechanical wear or assembly errors, or due to factors such as increased coil resistance, hysteresis and eddy current loss, etc. When the current loss increases, the electromagnetic attraction of the electromagnet may not be sufficient to pull the traction rod to complete the tripping action. At this time, the compensation current value calculated according to the TS model will increase accordingly. By increasing the compensation current, the electromagnetic attraction of the electromagnet is enhanced, thereby ensuring that the traction rod can be reliably pulled under the fault current and avoiding the release from refusing to operate. This compensation mechanism can effectively deal with mechanical and electrical uncertainties and improve the reliability of the system. The compensation circuit dynamically balances the electromagnetic attraction of the electromagnet with the elastic force of the reaction spring by adjusting the magnitude of the compensation current. Among them, the dynamic balancing mechanism ensures that the release can maintain stable mechanical properties under various working conditions. Under normal working conditions, the compensation current is maintained at an appropriate level, so that the electromagnetic attraction and the elastic force of the reaction spring offset each other, ensuring that the release will not malfunction. Under fault conditions, the compensation current can be quickly adjusted so that the electromagnetic attraction is sufficient to overcome the elastic force of the reaction spring and complete the tripping action. The compensation circuit outputs an equal value of compensation current based on the received compensation current value signal to accurately adjust the electromagnetic attraction of the electromagnet.

[0071] The output of the TS fuzzy controller specifically includes:

[0072] When the clear input quantity i is input to the TS-type fuzzy controller, the corresponding fuzzy rules are activated according to their membership, and the weight coefficient w of each rule is determined. Each activated fuzzy rule corresponds to a weight coefficient w, which reflects the degree to which the clear input quantity i belongs to the fuzzy set corresponding to the fuzzy rule. The TS-type fuzzy controller uses the weighted summation method or weighted average method to calculate the final output U. In the weighted summation method, the final output U is obtained by multiplying the clear output quantities u of all activated fuzzy rules with their corresponding weight coefficients w and then summing them. The calculation formula is: The final output U is obtained by multiplying the clear output u of all activated fuzzy rules by their corresponding weight coefficients w, and dividing it by the sum of all weight coefficients. The calculation formula is: , where u is the output of each rule and w is the corresponding weight coefficient;

[0073] The determination of the weight coefficient of each rule specifically includes:

[0074] The minimum method is used to determine the weight coefficient, that is, the minimum membership of each fuzzy rule is taken as the weight coefficient, wherein, for multiple fuzzy rules activated by a clear input quantity i, the weight coefficient w of each rule is set to the minimum membership value of its corresponding fuzzy set. By selecting the minimum membership, it is ensured that the activation degree of the rule will not be overestimated. The minimum method is suitable for scenarios with high requirements for system security and reliability, and can effectively avoid excessive control caused by excessive weights. The product method is used to determine the weight coefficient, that is, the product of the membership of each fuzzy rule is taken as the weight coefficient, wherein, for multiple fuzzy rules activated by a clear input quantity i, the weight coefficient w of each rule is set to the product of the membership of its corresponding fuzzy set, comprehensively considering all relevant The membership of the fuzzy set is concerned, and a more accurate weight coefficient is obtained through multiplication operation. The product method can better reflect the comprehensive membership of the input quantity in multiple fuzzy sets, and is suitable for scenarios that require precise control and comprehensive consideration of multiple factors. According to the actual application scenario and control requirements, the weight coefficient is artificially determined through human experience or experimental data to optimize the control effect. Among them, according to the specific needs and actual operation conditions of the system, the weight coefficient is artificially adjusted to optimize the control effect. It is necessary to combine the actual operation data, historical experience and expert knowledge of the system, and determine the most suitable weight coefficient through experimental verification and adjustment. It is suitable for complex systems or scenarios that require specific optimization goals, and can effectively improve the control performance and adaptability of the system.

[0075] The technical solution of the present invention is to implement a compensating circuit breaker release using a TS-type fuzzy controller. The basic principle of this technical solution is as follows: the use of a compensation circuit can effectively solve the problem of malfunction and refusal of the release, and the TS-type fuzzy inference output can be directly used to control the compensation circuit, which can approximate any nonlinear system. The fuzzy controller is a single-input-single-output system. By performing structural identification and parameter identification on current data, a TS model is established, and the TS model determines the working mode of the compensation circuit: when the elastic force of the reaction spring inside the electromagnet decays, the compensation current decreases, the electromagnetic attraction decreases, and the malfunction of the release is prevented; when the overlap distance between the iron core of the ferromagnet and the traction rod is too large, or the current loss increases, the compensation current increases, the electromagnetic attraction increases, and the refusal of the release is avoided. The TS-type fuzzy controller is used to control the current size of the compensation circuit of the circuit breaker, which can effectively prevent the malfunction and refusal of the release and prevent the short-circuit protection function of the circuit breaker from failing.

[0076] The compensation circuit also uses a compensation circuit based on a neural network, and its working methods include:

[0077] Receive the compensation current value signal from the fuzzy controller and adaptively adjust the compensation current through the compensation model built based on the neural network. The training data of the neural network includes the current signal under different working conditions and the corresponding compensation current value, and control the electromagnetic attraction of the electromagnet;

[0078] In addition, the construction process of the compensation model is:

[0079] The current signals and corresponding compensation current values ​​under different working conditions are collected, covering various situations such as normal working state, attenuation of reaction spring elastic force, excessive overlap distance between iron core and traction rod, and increased current loss. The collected data are normalized, the current signal and compensation current value are mapped to the same range, and the normalized data are integrated and divided into training set and test set, and then a neural network structure is designed, including input layer, hidden layer and output layer, wherein the number of neurons in the input layer is consistent with the number of features of the current signal, one or more hidden layers are designed, each layer contains several neurons, the hidden layer uses a nonlinear activation function, the number of neurons in the output layer is 1, representing the compensation current value, and a linear activation function is used to directly output the compensation current value. A compensation model is constructed based on the designed neural network structure, and the training set is used to input the neural network structure for training. The weights and bias of the neural network are randomly initialized, and the input current signal is passed through the neural network. Forward propagation, calculate the output compensation current value, use the mean square error as the loss function, calculate the difference between the compensation current value output by the neural network and the actual compensation current value, calculate the gradient of the loss function for each weight through the back propagation algorithm, and use the gradient descent method to update the weight to minimize the loss function, repeat the steps of forward propagation, loss calculation and back propagation until the loss function converges to a smaller value, and then use an independent test data set to verify the trained neural network, calculate the mean square error index on the test data, evaluate the performance of the neural network, and then optimize the neural network to finally obtain the constructed compensation model, deploy the constructed compensation model, use the compensation circuit to receive the compensation current value signal from the TS-type fuzzy controller, input the input signal into the compensation model, calculate the final compensation current value through forward propagation, and according to the output of the compensation model, the compensation circuit outputs the corresponding compensation current to adjust the electromagnetic attraction of the electromagnet;

[0080] The TS-type fuzzy controller also includes a fault diagnosis module, which is used to monitor the current signal of the circuit breaker and the working status of the compensation circuit in real time. When an abnormal signal is detected, the fault diagnosis module issues an alarm signal and transmits the abnormal information to the external monitoring system for fault detection and processing;

[0081] Among them, the specific working process of the fault diagnosis module is as follows: the fault diagnosis module continuously monitors the current signal of the line where the circuit breaker is located and the working status of the compensation circuit in real time, obtains the real-time value of the line current, and monitors the current, voltage and electromagnetic attraction data of the electromagnet in the compensation circuit, and performs real-time analysis and processing on the collected signals and data. It detects whether there is an abnormal signal through a preset threshold judgment. Among them, when the line current exceeds the set normal range threshold, or the parameters of the compensation circuit change that does not conform to the normal working mode, it is identified as an abnormal signal. Once an abnormal signal is detected, the fault diagnosis module immediately triggers the alarm mechanism and sends an alarm signal to alert on-site personnel. At the same time, the abnormal information including the specific parameters of the abnormal signal, the time of occurrence and the cause is transmitted to the external monitoring system through the communication interface. After receiving the abnormal information, the external monitoring system quickly locates the fault location and provides accurate fault information to the maintenance personnel so that the fault can be detected and processed in time, thereby minimizing the impact of the fault on the operation of the power system and ensuring the safe and stable operation of the power system.

[0082] Example 2, as Figure 2 As shown, based on Example 1, the present invention provides a technical solution: preferably, a TS-type fuzzy controller is used to implement a compensating circuit breaker release, and the compensation circuit of the circuit breaker release is a single-input-single-output system, and multiple sets of data of clear input quantity i and clear output quantity u are measured, such as Figure 2 As shown, from Figure 2 It can be seen that these test data are two-segment linear functions, and the slopes of the straight lines are different for different intervals of the input quantity i. Figure 2 The test data can be fitted to obtain the specific expression of the two-segment function and determine the coefficients a and k of the function. When a clear input i is measured, a clear output u can be obtained based on the two-segment function.

[0083] Example 3, as Figure 3 、 Figure 4 As shown, based on Examples 1-2, the present invention provides a technical solution: Preferably, based on a large amount of measured input-output data of a specific circuit breaker trip unit system, three TS-type fuzzy rules describing it are obtained by identification, and the membership functions are mf1, mf2 and mf3, respectively, as shown in FIG. Figure 3 、 Figure 4 As shown, at this time, if the current i1=2 in the system is measured, the activated rule is:

[0084] Fuzzy set mf1, weight coefficient is 0.8;

[0085] Fuzzy set mf2, weight coefficient is 0.4;

[0086] Using the weighted average method, the final total output is u=mf1*0.8+mf2*0.4;

[0087] Example 4, as Figure 3 、 Figure 5 As shown, based on embodiments 1-3, the present invention provides a technical solution: preferably, based on a large amount of measured input-output data of a specific circuit breaker trip unit system, three TS-type fuzzy rules describing it are obtained by identification, and the membership functions are mf1, mf2 and mf3 respectively, as shown in FIG. Figure 3 、 Figure 5 As shown, at this time, if the current i2=8 in the system is measured, the activated rule is:

[0088] Fuzzy set mf1, weight coefficient is 0.1;

[0089] Fuzzy set mf3, weight coefficient is 0.2;

[0090] Using the weighted average method, the final total output is u=mf1*0.1+mf3*0.2.

[0091] The present application effectively solves the problem of malfunction and refusal of the trip unit by adopting a compensation circuit, and the TS-type fuzzy reasoning output is used to control the compensation circuit to prevent the short-circuit protection function of the circuit breaker from failing.

[0092] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. Compensation type circuit breaker release based on TS type fuzzy controller, characterized by: The compensating circuit breaker tripper comprises an executive component, a compensation circuit and a fuzzy controller; Wherein, the actuator is an electromagnet, which is used to convert electromagnetic energy into mechanical energy to perform the tripping operation of the circuit breaker; The compensation circuit is connected to the electromagnet and is used to adjust the magnitude of the compensation current according to the control signal to adjust the electromagnetic attraction of the electromagnet; The fuzzy controller is a TS-type fuzzy controller, which is used to receive the current signal of the circuit breaker and output a control signal according to the preset fuzzy rules and TS model to control the working state of the compensation circuit, thereby preventing the trip unit from malfunctioning or refusing to operate; The compensation circuit also adopts a compensation circuit based on a neural network, and its working method includes: The compensation current value signal from the fuzzy controller is received, and the compensation current is adaptively adjusted through a compensation model constructed based on a neural network. The training data of the neural network includes the current signal and the corresponding compensation current value under different working conditions to control the electromagnetic attraction of the electromagnet.

2. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 1, characterized in that: The TS type fuzzy controller is a single-input-single-output system, and its control process includes the following steps: The current of the circuit breaker is monitored in real time, and a clear input quantity i is obtained through a current transformer. The clear input quantity i is the current value in the circuit breaker; The acquired current data is input into the TS-type fuzzy controller for structure identification and parameter identification; According to the identification result, a clear output quantity u is calculated and outputted through a preset fuzzy rule. The clear output quantity u is a compensation current value signal. The calculated compensation current value signal is then outputted to the compensation circuit. The compensation circuit receives the compensation current value signal from the TS type fuzzy controller, outputs the compensation current of equal value according to the compensation current value signal, and dynamically adjusts the electromagnetic attraction of the electromagnet by adjusting the magnitude of the compensation current.

3. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 2, characterized in that: The structure identification and parameter identification specifically include: Perform structural identification on the data of the clear input quantity i of the TS-type fuzzy controller to determine whether to use the 0th-order or 1st-order linear model; After determining the form of the linear model, further parameter identification is performed on the input current data according to the selected linear model to determine the parameters a and k in the model. For the 0th order model, the goal of parameter identification is to determine the constant k. For the 1st order model, both constants a and k need to be determined simultaneously. According to the results of structure identification and parameter identification, a complete TS model is established, and a clear output quantity u is output according to the model.

4. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 3, characterized in that: The control rules of the TS type fuzzy controller specifically include: Define the fuzzy set A, and determine the corresponding fuzzy rule according to the fuzzy set A to which the clear input quantity i belongs; In the 0th order model, the fuzzy rule is "if the clear input quantity i belongs to A, then u = k", where k is a constant related to the set A; In the first-order model, the fuzzy rule is "if the clear input quantity i belongs to A, then u = ai + k", where a and k are constants related to the set A; The corresponding fuzzy rules are activated according to the membership degree of the clear input quantity i, and the clear output quantity u is calculated according to the rules.

5. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 2, characterized in that: The working mode of the compensation circuit specifically includes: When the reaction spring force inside the electromagnet decays, the compensation current calculated according to the TS model decreases, and the electromagnetic attraction decreases accordingly to prevent the release from malfunctioning. When the distance between the ferromagnetic core of the electromagnet and the traction rod is too large, or the current loss increases, the compensation current calculated according to the TS model increases, and the electromagnetic attraction increases accordingly to avoid the release from refusing to operate; The compensation circuit dynamically balances the electromagnetic attraction of the electromagnet and the elastic force of the reaction spring by adjusting the magnitude of the compensation current; The compensation circuit outputs a compensation current of equal value according to the received compensation current value signal, thereby accurately adjusting the electromagnetic attraction force of the electromagnet.

6. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 5, characterized in that: The output of the TS type fuzzy controller specifically includes: When the clear input i is input into the TS fuzzy controller, the corresponding fuzzy rules are activated according to their membership, and the weight coefficient w of each rule is determined, where each activated fuzzy rule corresponds to a weight coefficient w; The TS-type fuzzy controller uses the weighted summation method or weighted average method to calculate the final output U; In the weighted summation method, the final output U is obtained by multiplying the clear output u of all activated fuzzy rules with their corresponding weight coefficients w and then summing them up. The calculation formula is: ; The final output U is obtained by multiplying the sum of the clear output u of all activated fuzzy rules with their corresponding weight coefficients w, divided by the sum of all weight coefficients. The calculation formula is: , where u is the output of each rule and w is the corresponding weight coefficient.

7. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 6, characterized in that: The determination of the weight coefficient of each rule specifically includes: The minimum method is used to determine the weight coefficient, that is, the minimum membership degree of each fuzzy rule is taken as the weight coefficient; The weight coefficient is determined by the product method, that is, the product of the membership degree of each fuzzy rule is taken as the weight coefficient; The weight coefficient is determined manually according to the actual application scenario and control requirements.

8. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 1, characterized in that: The construction process of the compensation model is as follows: The current signals and corresponding compensation current values ​​under different operating conditions are collected, covering various situations such as normal working conditions, attenuation of the reaction spring force, excessive overlap between the core and the drawbar, and increased current loss. The collected data is normalized and integrated, divided into training and test sets, and then the neural network structure, including the input layer, hidden layer, and output layer, is designed. A compensation model is constructed based on the designed neural network structure. The training set is used as input into the neural network structure for training. The weights and biases of the neural network are randomly initialized. The input current signal is forward-propagated through the neural network, and the output compensation current value is calculated. The trained neural network is then verified using an independent test data set. The mean square error index on the test data is calculated to evaluate the performance of the neural network. The neural network is then optimized to ultimately obtain the constructed compensation model. The constructed compensation model is deployed and the compensation circuit is used to receive the compensation current value signal from the TS-type fuzzy controller. The input signal is input into the compensation model, and the final compensation current value is calculated through forward propagation. According to the output of the compensation model, the compensation circuit outputs the corresponding compensation current to adjust the electromagnetic attraction of the electromagnet.

9. The compensating circuit breaker release based on the TS-type fuzzy controller according to claim 1, characterized in that: The TS-type fuzzy controller also includes a fault diagnosis module for real-time monitoring of the current signal of the circuit breaker and the working status of the compensation circuit. When an abnormal signal is detected, the fault diagnosis module issues an alarm signal and transmits the abnormal information to the external monitoring system for fault detection and processing; The specific working process of the fault diagnosis module is as follows: The fault diagnosis module continuously monitors the current signal of the circuit breaker and the working status of the compensation circuit in real time, obtains the real-time value of the line current, and monitors the current, voltage and electromagnetic attraction data of the electromagnet in the compensation circuit; The collected signals and data are analyzed and processed in real time, and the presence of abnormal signals is detected through preset threshold judgment. When the line current exceeds the set normal range threshold, or the parameters of the compensation circuit show changes that do not conform to the normal working mode, it is identified as an abnormal signal; Once an abnormal signal is detected, the fault diagnosis module immediately triggers the alarm mechanism and sends out an alarm signal to alert on-site personnel.

Images